Cerebrospinal fluid lactic acidosis bacterial meningitis

Archives of Disease in Childhood, 1981, 56, 692-698

Cerebrospinal fluid lactic acidosis in bacterialmeningitis

J EROSS, M SILINK, AND D DORMAN

Department of Endocrinology and Department of Bacteriology, Royal Alexandra Hospital for Children,Camperdown, New South Wales, Australia

SUMMARY A rapid, microenzymatic method was used to measure cerebrospinal fluid lactate levels in205 children with suspected bacterial meningitis. Fifty children with normal CSF containing fewerthan 0 005 x 109/1 WBC, no segmented neutrophils, glucose 3-4 ± 0 8 mmol/l (61 2 ± 14 4mg/100 ml), and a protein of less than 0 30 g/l had CSF lactate levels below 2-0 mmol/l (18 mg/100 ml) (mean and standard deviation 1I3 ± 0 3 mmol/l (11 8 ± 2-7 mg/100 ml)). In 31 cases ofproved viral meningitis as with 58 cases of clinically diagnosed viral meningitis, levels were below3-8 mmol/l (34-5 mg/100 ml), being 2-3 ± 0-6 mmol/l (20-9 ± 5.4 mg/100 ml), and 21 ± 0-7mmol/l (19 1 ± 6 4 mg/i 00 ml) respectively. Sixty-six cases of bacterial meningitis had CSF lactatelevels ranging from 3*9 mmol/l (35.4 mg/100 ml) to greater than 10 0 mmol/l (90.0 mg/100 ml).Longitudinal studies in 7 children with bacterial meningitis showed that cerebrospinal fluid lactatelevels differentiated bacterial from viral meningitis up to 4 days after starting treatment with anti-biotics. Use of CSF lactate measurement for monitoring the efficacy of treatment is illustrated in a

case of bacterial meningitis due to Pseudomonas aeruginosa. The origin of the cerebrospinal fluidlactate acidosis and the role of lactate in the pathophysiological cycle leading to intensification ofbrain tissue hypoxia and cellular damage is discussed with respect to the short-term prognosis andthe long-term neurological sequelae.

With the advent of antibiotics three decades ago themortality rate in bacterial meningitis was reducedfrom 50-900% to about 5-10%. There has been nofurther comparable improvement since then.'-3However studies on children with meningitis due toHaemophilus inflitenzae indicate that as many as 30%of the survivors suffer from severe neurological andpsychological sequelae.A5To date there is no laboratory test which is both

rapid and reliable for differentiating bacterialmeningitis from viral meningitis or for quantitatingthe effectiveness of treatment. Results from a carefulinspection of the cerebrospinal fluid (CSF) withcomplete differential cell count, and glucose andprotein levels may still leave the differential diagnosisin doubt.6 Gram stains may be negative in as many as30% of cases of culture-proved bacterial meningitis.7Furthermore, treatment with antibiotics beforeinitial lumbar puncture, which occurs in about halfthe cases of bacterial meningitis, 12 increases bothnegative Gram stains and negative cultures.10

Several methods have been suggested for the rapiddifferential diagnosis of meningitis-such as CSF

lactate dehydrogenase activity,13 counterimmuno-electrophoresis,'4 nitroblue tetrazolium test,15 thelimulus lysate test,'6 and CSF pH measurement.However some of these tests produced more than10% false-negative results, others detected onlysome types of organisms, and the equipment neededand the expertise required were beyond the scope ofmost clinical laboratories. Moreover there wasdecreased reliability if antibiotics had been given.

In 1917 Levinson17 observed low pH values in theCSF of patients with bacterial meningitis, andbelieved this was due to lactic acid. In 1925 Killian'8presented supporting data on CSF lactate from 5normal patients and 25 patients with bacterialmeningitis. In 1933 De Sanctis et al.19 concluded thatvariations in lactic acid of spinal fluid offered more-reliable information about the clinical progressduring treatment of bacterial meningitis than CSFsugar or total leucocyte counts.

Since Killian's report'8 numerous papers haveproduced results in support of the measurement ofCSF lactate levels in the differential diagnosis ofmeningitis.2032 Despite this, the use of CSF lactate

692

copyright. on July 21, 2022 by guest. P

rotected byhttp://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.56.9.692 on 1 S

eptember 1981. D

ownloaded from

http://adc.bmj.com/

Cerebrospinal fluid lactic acidosis in bacterial meningitis 693

measurements as a rapid screening test for bacterialmeningitis has not gained wide acceptance, primarilyowing to the technical difficulties of its measurement.

This paper presents the results of our experiencewith CSF lactate measurement in normal childrenand children with meningitis using a rapid enzymaticmicromethod slitable for a small clinical laboratorywith standard laboratory equipment.

Materials and methods

CSF samples were obtained from 207 children agedbetween 2 days and 15 years. Each had had a lumbarpuncture because of clinical features that suggestedbacterial meningitis.Complete differential cell count, a Gram stain,

glucose and protein determinations were performedon each CSF sample.No special preparation of the CSF was required

for lactate determination except immediate centri-fugation, using the supernatant if grossly blood-stained. If a sample could not immediately beassayed it was stored at a temperature of -200C.For this method33 lactate was converted to

pyruvate by the enzyme lactic dehydrogenase (LDHEC 1.1. 1.27) in a stoichiometric reaction with nicotin-amide adenine dinucleotide (NAD +). Pyruvateinhibition of LDH was removed by reaction of thepyruvate with either glutamate pyruvate trans-aminase (GPT EC 2.6.1.2) to form alanine at pH 8 - 9,or alternatively, with hydrazine to form the hydra-zone at pH 9 5. The increase in reduced nicotineadenine dinucleotide was measured fluorometricallyor spectrophotometrically.The assay protocol consisted of adding 2-10 pl of

the CSF sample to 1 *0 ml of reaction mixturecontaining 25 ml glutamate buffer (Sigma) to aconcentration of 0-120 mol/l, 60 ml fresh glassdistilled H20, 5 ml of NAD (Sigma grade III) to aconcentration of 0 80 mmol/l, and either 2 units/mlof GPT (Boehringer Mannheim from pig heart) or5 ml of hydrazine (BDH AR grade) to a concen-tration of 0 103 mol/l, made up to 100 ml withdistilled H20.

Samples were measured in duplicate in 10 x 75mmPyrex disposable tubes (Corning). Stock standard oflithium 1-lactate (10 mmol/l) was diluted with thereaction mixture to 1 mmol/l, and 0-50 ,l per tubecontaining 0-50 nmol/l was used to construct a

standard curve.Fourteen units of LDH from rabbit muscle

(Boehringer Mannheim) stored in tris buffer20 mmol/l, pH 8 0, and 0*02% bovine serum

albumin were added to each tube.The tubes were vortexed and incubated for 30

minutes at 370C, allowed to cool, and then read ineither a Perkin Elmer fluorimeter 650 lOS (onsensitivity 3 and slits 3 nm, excitation wavelength360 nm, analytical wavelength 465 nm) or a unicamSP 500 spectrophotometer at 365 nm.Each patient was placed in one of four groups.

Group 1 (n = 50). Normal, WBC< 0 005 x 109/lwith no segmented neutrophils, normal glucose andprotein, and a clear Gram stain. No cultures were

set up on these samples.Group 2 (n = 31). Viral meningitis, identified virusisolated from CSF.Group 3 (n = 66). Bacterial meningitis, positiveculture with identification of the organism.Group 4 (n = 58). Miscellaneous, leucocytosis ofunknown aetiology. These were presumed to bepatients with viral meningitis from whose CSF no

bacteria or virus could be isolated.

Results

CSF results for the four groups are given in Table 1.The normal children all had CSF lactate levels less

than 2v0 mmol/l (18.0 mg/100 ml) (mean andstandard deviation 1 * 3 0 3 mmol/l (11 8 ± 2 7mg/100 ml)). Children with proved or suspectedviral meningitis had CSF lactate levels less than3-8 mmol/l (34-5 mg/100 ml)-2 3 ± 0 6 and2-1 I 0-7 mmol/l (20-9 ± 5*4 and 19 1 i 6-4mg/100 ml) respectively. Children with bacterialmeningitis had CSF lactate levels from 3-9 mmol/l(35-4 mg/100 ml) to greater than 10 0 mmol/l

Table 1 CSF measurements (means and SD) made in 205 children

Group Median age (years) WBC (x 106/i) % segmented Glucose (mmol/l) Protein (gll) Lactate (mmol/l)neutrophils

Normal (n = 50) 3.8 <5 0 34 +0-8 <0-30 1-3 0-3Viral (n = 31) 5-0 335 37 3-3 +08 0-10-5-50 2-3 0-6

(23 - 1250) (0 - 100)Bacterial (n = 66) 1-1 8227 81 <1.0 -3.8 0-15- >10.0 3-9 >10.0

(125 - 50000) (2 - 100)Others (n = 58) 3.7 572 31 3.2 + 0.7 0.10 - 10.0 2.1 ±07

(100 -3600) (0 - 100)

Conversion: SI to traditional units: Lactate 1 mmol/I1 9-0 mg/100 ml, glucose: 1 mmol/l ; 18 mg/100 ml.Figures in brackets are ranges.



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/

694 Eross, Silink, and Dorman

Normal Provedviral

(n=50) (n=31)

Proved Presumedbacterial viral(n=66) (n=58)

Table 2 Causative agents in groups 2 and 3

Group 2Viral meningitisMumps virusEcho 30Coxsackie B4Coxsackie B3Echo 1 1Coxsackie B2

Group 3Bacterial meningitis

Haemophilus influenzaeStreptococcus pneumoniaeNeisseria meningitidisEscherichia coliGroup B streptococciPseudomonas aeruginosaKlebsiella aerogenesCitrobacter diversus

No

156432I

30141342

l

I

Fig. 2 shows the effect of antibiotics on lactatelevels in 7 children with bacterial meningitis. Threehad infections due to Escherichia sp., one withH. influenzae, one with Pseudomonas aeruginosa, onewith Citrobacter diversus, and one with N. menin-gitis group A. In each case the organism wassensitive to the antibiotic that was given parenterallyin appropriate doses.

Fig. 1 CSF lactate levels in 205 children. Each dotrepresents levelfound in one child.

(90.0 mg/100 ml). Individual levels from the 205children are shown in Fig. 1.Two false-negative results were obtained. One was

from a child who had meningococcal septicaemiawhen an initial lumbar puncture showed WBC0-002 x 109/l, 1 being a polymorph, RBC 0-014 x109/l, and a protein concentration of less than0 10 g/l; glucose was 4-1 mmol/l (73-8 mg/100 ml),lactate 2 4 mmol/l (21 -8 mg/100 ml), and the Gramstain was clear. However a culture grew Neisseriameningitidis. Because the child's condition rapidlydeteriorated a second lumbar puncture was per-formed 8 hours later; this showed WBC 8 85 x 109/l,98% polymorphs, protein 2- 5 g/l, glucose 1 * 5 mmol/l(27 mg/100 ml), and lactate 10 0 mmol/l (90 mg/100 ml).The other false-negative result was also from a

child with septicaemia, the lactate changing from2-2 mmol/l (20 mg/100 ml) to 15-4 mmol/l (140 mg/100 ml) in 12 hours.

Thirty-nine per cent of patients in the bacterialgroup had had treatment with antibiotics.A breakdown of the various agents responsible

for meningitis in the viral and bacterial groups isgiven in Table 2.

8

7.

6.

5 -

Lactate(mmol/l) 4

3-

2-

1 1

0-4 5-9Days of treatment

10

8

7

Lactate(mmol/l )

6

5,~

3,

2

1'

0

..::...:::::....::11-

II@

Um UA A

U

*0 U0

A

V4K*A

A

0

A

0~~~~

*

_ - - * , --10-14 15

Fig. 2 Effect of treatment with antibiotics on CSFlactate in 7 children with bacterial meningitis. Levelsmeasured in any one child on different days are shownby same symbol.

Il

. .

. .



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/


Case 1 age lmonthGrowth [Pseudomonas aeruginosa[43601

Insertion of -Removal ofR ickham reservoir Growth Rickhom reservoir

[941

1841 ~~~~~~~~~~~~~~~[351

1 .i1 3 5 7 9

Carbenicitlin 250mg 6/24hours

5 Gentomicin 12/24 hours IVImg

11 13 15Days

17 19 21 23 25

t t t t t2mg intraventricular gentamicin

t t. 3mg intrathecal <-

Fig. 3 CSF lactate profile in a patient with meningitis due to Pseudomonas aeruginosa. Figures in brackets are

WBC x 106. Dotted line represents upper linmit to the normal range.

An individual response of lactate to antibiotictreatment of bacterial meningitis is illustrated inFig. 3. This patient, a 1-month-old boy (Case 1), wasadmitted with a history of persistent central nervous

system infection due to P. aeruginosa that did notrespond well to 21 days of parenteral gentamicintreatment. An initial CSF sample on admissionshowed WBC 4-3 x 109/l of which 85% were

polymorphs, protein greater than 10 g/l, glucose2-0 mmol/l (180 mg/100 ml), and lactate 70 mmol/l(63 * 6 mg/100 ml). A culture of the CSF subsequentlygrew P. aeruginosa. Treatment with intravenousgentamicin (4 mg twice daily) and carbenicillin(250 mg four times daily) was started on day 1,followed by intraventricular gentamicin (2 mg daily)on day 3 after insertion of a Rickham reservoir.

His clinical condition improved with a rapid fall inleucocytes in his CSF, and the lactate level was

2-0 mmol/l (18 0 mg/100 ml) on day 5 whencarbenicillin was stopped because the organism wasshown to be resistant to it. Intraventricular genta-micin was also stopped and on day 8 the dose ofintravenous gentamicin dose was halved. Later thatday he became febrile, a CSF sample on day 10showed WBC 1-54 x 109/l of which 75% werepolymorphs, protein greater than 10 g/l, glucose

1-9 mmol/l (34 2 mg/l00 ml), and a lactate level of5-1 mmol/l (46.3 mg/100 ml). A culture of thissample again showed growth of P. aeruginosa.Treatment with gentamicin was continued

systemically, intraventricularly, and, finally, intra-thecally (after removal of the Rickham reservoir on

day 17). There was a progressive clinical improve-ment with a concomitant fall in the level of CSFlactate.

Discussion

The fact that CSF lactate levels overlap between thenormal and viral groups has been observed by others(Table 3). In the present series CSF lactate levelsWere higher than in viral meningitis in all cases ofbacterial meningitis with the exception of two. These2 false-negatives show that the CSF lactate level is ameasure of hypoxic brain damage rather than an

indicator of organisms. Theoretically there must besome point at which an organism responsible for thesepticaemia crosses the blood brain barrier. While itis present in the CSF of the child with meningitisthere would be a time lag before any appreciablecentral nervous system damage takes place and thelactate level becomes increased. These results show

6 5

E

E

4.a

-a3

2

u Li



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/


Table 3 CSF lactate levels taken from other reports

Reference Controls Meningitis Method(mmol/l)

No Date Viral Bacterial(mmol/l) (mmol/l)

18 1925 0.9- 1.7 - 1.2-8 5 Colorimetric(n=5) (n=17)

19 1933 1.3-1 8 - 3-0-19-6 Colorimetric(n =27)

20 1964 1-6+0-4 2-0±0-2 6-6-21.5 Enzymatic(n=23) (n =5) (n= 10)

21 1970 1*6+02* 2-3 +0*8 6*7+41*8 Colorimetric(n =20) (n= 17) (n= 19)

22 1972 1-7+0-6 2-1+0-7 7.22-+1 Enzymatic(n=46) (n 35) (n=21)

23 1974 1-6+0-1 2-2±1 83+0-7 Enzymatic(n =25) (n =25) (n= 15)

24 1977 1-5+0-7 1-5+0-7 6-0+1-9 Enzymatic(n=20) (n =36) (n= 16)

26 1977 0.5 -8.4 1-1-5-0 8-0+28-5 GLC(n= 106) (n= 15) (n =8)

28 1977 1*3±0*6 2-1±07 6-4+3*6 Colorimetric(n= 10) (n= 10) (n=22)

30 1978 1-6+0-8 2-3±1-0 6-5+1-0 GLC(n =25) (n =26) (n=4)

This report 1 3+0-3 2-340.6 3.9-> 10.0 Enzymatic(n= 50) (n =31) (n= 66)

that this time lag is less than 8 hours and may in factonly be a matter of minutes.Thus CSF lactate levels could be used to differen-

tiate bacterial from viral meningitis in most of our

cases. Some studies have described such an overlapespecially in cases of tuberculous meningitis.24 28 32We had only one child who was clinically diagnosedas having tuberculous meningitis but in whom no

organism could be cultured; the CSF lactate was

3.5 mmol/l (31-8 mg/100 ml). Brook et al.30 haveshown that there is no age or sex difference in CSFlactate levels regardless of the aetiological group.

All CSF measurements (Table 1) were similar forthe viral and miscellaneous groups, as would beexpected.The CSF lactate levels in the 7 treated patients with

bacterial meningitis (Fig. 2) showed that suchchildren could still be distinguished from childrenwith viral meningitis up to 4 days after startingantibiotics.The decrease in lactate levels after initiation of

antibiotic treatment in the 7 patients agrees withother studies.18 19 21 22 24 28 The time for the lactateconcentration to reach viral levels in these reportsvaried from 1 to 5 days. It is evident from our resultsas well as those from other reports that each childresponds in a different manner. Brook et al.30showed that the rate of decrease was dependent on

the type of organism, tuberculous meningitisrequiring more than 6 weeks before lactate levelsreached normal levels. Since lactate is cleared onlyslowly from the CSF by diffusion into the brain

tissue,34 35 the lactate level before starting treatmentand the 'aggressiveness' of treatment would be twoadditional factors.The lactate profile in Case 1 (Fig. 3) illustrates the

potential use of CSF lactate for monitoring theeffectiveness of treatment in bacterial meningitis. Theinitial rapid fall in CSF lactate levels from 7 - 0 mmol/l(63-6 mg/100 ml) on day I to 2-0 mmol/l (18-0 mg/100 ml) on day 5 took place when the leucocytecount was falling, cultures were negative, and therewere signs of clinical improvement. Synergismbetween carbenicillin and gentamicin36 together withcombined intravenous and intraventricular genta-micin might have been responsible for this rapidimprovement. The fact that carbenicillin was stoppedon this day, and the intravenous dose of gentamicinhalved on day 8 (at a time when the penetration ofgentamicin through the blood brain barrier mighthave been decreasing37) and the fact that intra-ventricular gentamicin was given only sporadicallywould explain the reisolation of the organism on day10. The lactate rose to 5 1 mmol/l (46* 3 mg/100 ml)during this period demonstrating that the biochemicaldeterioration was parallel to the clinical condition.Intravenous, intraventricular, and intrathecal treat-ment with gentamicin was continued and thisproduced asepsis associated with virtually normallactate levels.

Despite many suggestions, the cause of the lacticacidosis in bacterial meningitis is still not clear.8Montani and Perret20 claimed that the CSF lactatelevel was dependent on two factors-the number oforganisms and the number of leucocytes present. Therelationship to the number of leucocytes is debatablefor a number of reasons. Firstly only a small amountof lactate is produced by endotoxin-stimulatedleucocytes,3840 increased CSF lactate occurs intuberculous meningitis in patients with few bacteriaand a mononuclear cellular response,41 and CSFlactic acidosis can be found in severe cases of acuteencephalitis in the presence of a low leucocyte count.Hansen et al.42 suggested the increase in lactate wasdue partly to inflamed brain tissue and partly todecerebrate rigidity and convulsions. Simpson et al.43demonstrated moderate increases in CSF lactate in 7of 22 patients with prolonged seizures (longer than30 minutes) or recurrent short convulsions.Kopetzky and Fishberg44 suggested that the lactic

acidosis in purulent meningitis might be due to adecrease in cerebral blood supply caused by increasedintracranial pressure. In fact a moderate increase inintracranial pressure will appreciably reduce cerebralperfusion pressure thereby reducing cerebral bloodflow, shifting brain metabolism from aerobic toanaerobic glycolysis, with increased lactate levels in



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/


brain tissue and CSF.45 46 Raisis et al.47 havedemonstrated the dependence of cellular oxidativemetabolism on the maintenance of adequateperfusion pressure in hydrocephalus, and Paulsonet al.48 found that cerebral blood flow was decreasedby as much as 40% in pyogenic meningitis, whileCSF lactate was increased, CSF bicarbonatereduced, and autoregulation often impaired.CSF hydrogen ion concentration and lactate are

concerned in a cycle of pathophysiological mechan-isms (Fig. 4) involving vasoparalysis with the loss ofvasomotor autoregulation, CO2 regulation of vaso-motor tone, resulting in the spreading of initiallylocalised oedema.49 Thus CSF lactic acidosis isdangerous because of its intensification of tissuehypoxia.

It is worth noting that several studies5054 onconditions other than infection indicate that CSFlactate levels below 3 0 mmol/l (27 mg/100 ml) areneurologically 'safe' while values above 4 0 mmol/l(36 mg/100 ml) are found in life-threatening states inwhich survival could lead to permanent neurologicaldamage. This suggests the prognostic value of CSFlactate in determining cellular damage, and mayprove to correlate with the long-term neurologicaland psychological deficits observed in childrensurviving bacterial meningitis.

Seitz and Ocker,54 in an attempt to break thevicious circle (Fig. 4) that leads to non-reversiblehypoxia in the brain tissue of severely brain-injuredpatients with CSF lactic acidosis, used a solution of8 - 4% sodium bicarbonate administered intrathecally.This treatment resulted in an increase in overallcerebral blood flow of 33 %, with 60% of the patientsshowing an improvement in clinical condition.

Local brainoedema

VasoparalysisIncreased intracranial pressure

Vasoparalysis Decreased regionallass at autoregulation cerebral blood flowand CO2ra

Local hypoxia;lactic acidosis

Fig. 4 Pathophysiology of CSF lactic acidosis.

The rehabilitation of such patients improved and themortality rate was reduced.

This study confirms the value of the estimation ofCSF lactate as a diagnostic aid in meningitis. It alsoindicates the potential of lactate as a biochemicalmeasurement for monitoring the efficacy oftreatmentin bacterial meningitis. With respect to short-termprognosis and the long-term neurological sequelae,the possible use of intrathecal sodium bicarbonate inthe treatment of severe brain injury with life-threatening CSF lactic acidosis would seem worthyof investigation in bacterial meningitis.

We thank Dr H Kilham and Dr H Kitson forconstructive criticism and Lina Pipitone forsecretarial expertise.

References

1Public Health Service: Epidemiological ResearchLaboratory. Deaths from bacterial meningitis. Br Med J1975; iii: 714-5.

2 Nyhan W L, Richardson F. Complications of meningitis.Annu Rev Med 1963; 14: 243-60.

3Sproles E T, III, Azerrad J, Williamson C, Merrill R E.Meningitis due to Hemophilus influenzae: long termsequelae. J Pediatr 1969; 75: 82-8.

4Lindberg J, Rosenhall U, Nylen 0, Ring& A. Long termoutcome of Hemophilus influenzae meningitis related toantibiotic treatment. Pediatrics 1977; 60: 1-13.Sell S H W, Merrill R E, Doyne E 0, Zimsky E P, Jr.Long term sequelae of Hemophilus influenzae meningitis.Pediatrics 1972; 49: 206-11.

6 Feigin R D, Shackelford P G. Value of repeat lumbarpuncture in differential diagnosis of meningitis. N Engl JMed 1973; 289: 571-4.

7Hyslop N E, Swartz M N. Bacterial meningitis. PostgradMed 1975; 58: 120-8.

8 Harter D H. Preliminary antibiotic therapy in bacterialmeningitis. Arch Neurol 1963; 9: 343-7.Dalton H P, Allison M J. Modification of laboratoryresults by partial treatment of bacterial meningitis. Am JClin Pathol 1968; 49: 410-3.

10 Jarvis C W, Saxena K M. Does prior antibiotic treatmenthamper the diagnosis of acute bacterial meningitis? ClinPediatr (Phila) 1972; 11: 201-4.

'1 Converse G M, Gwaltney J M, Jr, Strassburg D A,Hendley J 0. Alteration of cerebrospinal fluid findings bypartial treatment of bacterial meningitis. J Pediatr 1973;83: 220-5.

12 Quaade F, Kristensen K P. Purulent meningitis: a reviewof 658 cases. Acta Med Scand 1962; 171: 543-50.

13 Neches W, Platt M. Cerebrospinal fluid LDH in 287children, including 53 cases of meningitis of bacterial andnon-bacterial etiology. Pediatrics 1968; 41: 1097-103.

14 Coonrod J D, Rytel M W. Determination of aetiology ofbacterial meningitis by counterimmunoelectrophoresis.Lancet 1972; i: 1154-7.

15 Fikrig S M, Berkovich S, Emmett S M, Gordon C. Nitro-blue tetrazolium dye test and differential diagnosis ofmeningitis. J Pediatr 1973; 82: 855-62.Nachum R, Lipsey A, Siegel S E. Rapid detection ofGram-negative bacterial meningitis by the limulus lysatetest. N Engl J Med 1973; 289: 931-7.



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/


17 Levinson A. The hydrogen ion concentration of cerebro-spinal fluid: studies in meningitis. J Infect Dis 1917; 21:556-70.

18 Killian J A. Lactic acid of normal and pathologicalspinal fluids. Proc Exp Biol Med 1925; 23: 255-7.

19 De Sanctis A G, Killian J A, Garcia T. Lactic acid ofspinal fluid in meningitis. Am J Dis Child 1933; 46:239-49.

20 Montani S, Perret C. Acidose lactique du liquide cephalo-rachidien dans les meningites bacteriennes. Schweiz MedWochenschr 1964; 44: 1552-7.

21 Yalaz K. Cerebrospinal fluid lactate concentration inneurologically normal and abnormal children. Turk JPediatr 1970; 12: 8-15.

22 Lamisse F, Grenier B, Choutet P, Rolland J C, Gautier J.Interet du dosage du l'acide lactique dans le liquiderachidien pour le diagnostic des meningites purulentes.Lyon Mid 1972; 228: 591-5.

23 Bland R D, Lister R C, Ries J P. Cerebrospinal fluid,lactic acid level, and pH in meningitis. Am J Dis Child1974; 128: 151-6.

24 Fiero H, Chacon A, Gutierrez G, Munoz 0. Valores depH y acido lactico en el liquidae cefalerraquideo comoayuada diagn6stica para differenciar meningeoencefalitisviral y purulenta. Arch Invest Med (Mex) 1977; 8: 185-92.

25 Controni G, Rodriguez W J, Hicks J M, et al. Cerebro-spinal fluid lactic acid levels in meningitis. J Pediatr 1977;91: 379-84.

26 Ferguson I R, Tearle P V. Gas liquid chromatography inthe rapid diagnosis of meningitis. J Clin Pathol 1977; 30:1163-7.

27 Lauwers S. Apport du dosage d'acide lactique dans leLCR au diagnostic differentiel des meningites. Acta ClinBelg 1977; 32: 319-25.

28 Gandhi P C, Khan M A, Mathur P S, Singh V. Lacticacid levels in cerebrospinal fluid in various meningitidesin infancy and childhood. Indian Pediatr 1977; 14: 505-10.

29 Hurd C A, Fennema K J, Blazevic D J. Use of cerebro-spinal fluid lactate levels in the diagnosis of bacterialmeningitis. Am J Med Technol 1978; 44: 11-3.

30 Brook I, Bricknell K S, Overturf G D, Finegold S M.Measurement of lactic acid in cerebrospinal fluid ofpatients with infections of the central nervous system.JInfect Dis 1978; 137: 384-90.

31 Komorowski R A, Farmer S G, Hanson G A, Hause L L.Cerebrospinal fluid lactate acid in diagnosis of meningitis.J Clin Microbiol 1978; 8: 89-92.

32 D'Souza E, Mandal B K, Hooper J, Parker L. Letter:Lactic acid concentration in cerebrospinal fluid anddifferential diagnosis of meningitis. Lancet 1978; ii:579-80.

33 Noll F. Determination of L-lactate with LDH, GPT, andNAD. In: Bergmeyer H U, ed. Methods of enzymaticanalysis. Vol. 3. London: Verlag Chemie Weinheim/Academic Press, 1974: 1475-9.

34 Prockop L D. Cerebrospinal fluid lactic acid clearanceand effect on facilitated diffusion of a glucose analogue.Neurology (Minneap) 1968; 18: 189-96.

35 Valenca L M, Shannon D C, Kazemi H. Clearance oflactate from the cerebrospinal fluid. Neurology (Minneap)1971; 21: 615-20.

36 Holmes R K, Minshew B H, Sanford P J. Resistance ofPseudomonas aeruginosa to aminoglycoside antibiotics.Jlnfect Dis 1974; 130: S163-6.

37 Smith D H, Ingram D L, Smith A L, Gilles F, BresnanM J. Bacterial meningitis: a symposium. Pediatrics 1973;52: 586-600.

38 Menkes J H. The causes for low spinal fluid sugar inbacterial meningitis: another look. Pediatrics 1969; 44:1-3.

39 Cohn Z A, Morse S I. Functional and metabolic proper-ties of polymorphonuclear leucocytes. II. The influence ofa lipopolysaccharide endotoxin. J Exp Med 1960; 111:689-704.

40 Himwich W A, Himwich H E. Pyruvic acid exchange ofthe brain. J Neurophysiol 1964; 9: 133-6.

41 Sudre Y, Reiss D. Interet du dosage de l'acide du liquidecephalo-rachidien dans les meningites. Bordeaux Med1969; 2: 2045-57.

42 Hansen E L, Kristensen H S, Brodersen P, Paulson 0 B,Miillertz S, Jessen 0. Acid base pattern of cerebrospinalfluid and arterial blood in bacterial meningitis and inencephalitis. Acta Med Scand 1974; 196: 431-7.

43 Simpson H, Habel A H, George E L. Cerebrospinal fluidacid-base status and lactate and pyruvate concentrationsafter convulsions of varied duration and aetiology inchildren. Arch Dis Child 1977; 52: 844-9.

44 Kopetzky S J, Fishberg E H. Changes in distributionratio ofconstituents ofblood and spinal fluid in meningitis.JLab Clin Med 1933; 18: 796-801.

45 Langfitt T W. Increased intracranial pressure. ClinNeurosurg 1968; 16: 436-71.

46 Zwetnow N N. Effects of increased cerebrospinal fluidpressure on the blood flow and on the energy metabolismof the brain. An experimental study. Acta Physiol Scand[Suppl] 1970; Supplement 339,1-28.

47 Raisis J E, Kindt G W, McGillicuddy J E, Miller C A.Cerebrospinal fluid lactate and lactate/pyruvate ratios inhydrocephalus. JNeurosurg 1976; 44: 337-41.

48 Paulson 0 B, Brodersen P, Hansen E L, Kristensen H S.Regional cerebral blood flow cerebral metabolic rate ofoxygen and cerebrospinal fluid acid-base variables inpatients with acute meningitis and with acute encephalitis.Acta Med Scand 1974; 196: 191-8.

49 Reulen H J, Hadjidimos A, Hase U. Steroids in thetreatment of brain edema. Adv Neurosurg 1973; 1:92-105.

50 Crockard H A, Taylor A R. Serial CSF lactate/pyruvatevalues as a guide to prognosis in head injury coma. EurNeurol 1972; 8: 151-7.

51 Czernicki Z, Grochowski W. Cerebrospinal fluid lactatesfollowing craniocerebral injuries (in Polish). NeurolNeurochir Pol 1976; 10: 651-3.

52 Raudam E 1, Rivis E K. The prognostic value of thelactate concentration and the lactate/pyruvate ratio incerebrospinal fluid following acute cerebral circulatorydisorders (in Russian). Vopr Neirokhir 1976; 2: 32-7.

53 Cold G, Enevoldsen E, Malmros R. Ventricular fluidlactate, pyruvate, bicarbonate, and pH in unconsciousbrain-injured patients subjected to controlled ventilation.Acta Neurol Scand 1975; 52: 187-95.

54 Seitz H D, Ocker K. The prognostic and therapeuticimportance of changes in the CSF during the acute stageof brain injury. Acta ANeurochir (Wien) 1977; 39: 211-31.

Correspondence to Mr J Eross, Department ofEndocrinology, Royal Alexandra Hospital forChildren, Camperdown, Sydney, New South Wales2050, Australia.

Received 9 June 1980



j.com/

Arch D


eptember 1981. D

ownloaded from

http://adc.bmj.com/

Documents

Cerebrospinal fluid lactic acidosis bacterial meningitis